The Piano as Orchestra, Part 1 of ?

I wrote yesterday about wishing that Copland had created a better piano reduction for his Clarinet Concerto. It's especially frustrating because the music lends itself well to piano - piano is even prominently featured in the original orchestration; however, the version he created looks more like a study score. The other two works I'm playing on tonight's program, Debussy's Premiere Rhapsody and Lutoslawski's Dance Preludes, were originally written for piano accompaniment, although each was later orchestrated by its composer. Curiously, as wonderfully as Debussy usually writes for the piano, I don't think this is his most idiomatic piano writing either - I would have guessed it was an orchestral reduction if I didn't know otherwise.

I've always been fascinated by piano reductions. Back when I was languishing in ABDdom, I had the ongoing, though unwieldy idea of writing something called "The Piano as Orchestra." (The subtitle would have to be "Fake Your Tutti.") I still hope to tackle it some day, although the book I'd like to write would have a much broader focus than a dissertation. Basically, it would be a way of looking at all of my favorite music & meaning topics: translation, identity, faking, constraints. More specifically, it would allow me to explore one of my favorite musical activities - the sometimes Walter Mitty-like experience of pretending that 10 fingers and 88 keys can do what a full orchestra can do.

One of my favorite iPod playlists is full of various tracks that fall under this "piano as orchestra" umbrella: Beethoven symphonies arranged by Liszt; many other works, including The Rite of Spring, Prelude to the Afternoon of a Faun and Bolero (!) arranged for two pianos; Alkan's unbelievable Concerto for Solo Piano, meant to imply piano vs. orchestra (I have to write about this piece soon); Glenn Gould's astonishing La Valse, etc. It's not that I don't like orchestral color, but sometimes the B&W version provides a nice change of pace. When Strauss orchestrated his most perfect song, "Morgen," he logically handed the melody from the original piano accompaniment to a solo violin. And yet, to me, the effect is sickly sweet and no more satisfying than a colorized version of a film classic. The way that melody floats in the piano version can't be improved. (Although it's an interesting philosophical question to wonder how I'd feel if the orchestra version had come first.)

I'll close with an example that is both clearly not an ideal transcription and that also reminds me how much fun I have faking my way through a reduction without really trying to be perfect. I wish I could improvise like a jazz musician, but this video kind of illustrates what improvisation means to me, a good sight-reader who doesn't always like to practice. For my daughter' 4th birthday party several years ago, we decided to have a puppet party in which the children would make little stick puppets. Somehow, at the eleventh hour we decided also to put on a puppet show of Peter and the Wolf; my wife, my sister and I constructed the puppets the night before and then performed an abridged version of Prokofiev,with the two of them working the puppets and me at the piano.

Now given that this work is designed to show off the varied colors of an orchestra, it's obviously odd to pianocize it, but Prokofiev's music actually holds up pretty well that way. The tunes themselves are so good and full of character, and Prokofiev's music is just naturally pianistic. I'd like to emphasize that I did not practice much, so not every note goes where it's supposed to go, but I did have fun. The version below was actually shot after the party with just my daughter as the audience - we wanted one more crack at it, although there are still some technical snafus, especially with the annoying rope that appeared too early. (I tried to edit that out in the video, not always successfully.) Unfortunately, the balance is a little heavy towards the piano, so the narration gets drowned out occasionally, but you probably know the basics of the story. Honestly, I enjoy this as much for the amusement provided by the puppets and the puppetry as by Prokofiev and the piano, but polished it's not.

Imitation as Inspired Improve-ization

As I recall from the 30 seconds I just spent researching it, Plato believed that "art is imitation, and that's bad," and Aristotle believed that "art is imitation, and that's all right, even good." Well, I'm here to say that art is imitation, and that in some cases, the imitation creates something better than the original - in part because it fails as imitation.

Consider the third of Bartók's Romanian Folk Dances, a little set of six he pieces he wrote for piano.

Against a simple left hand drone, the right hand plays an exotic little melody which you can hear here (played by Bartók!) at the 1:23 mark:

As it happens, just yesterday on the radio I heard a performance of the composer's orchestration of these dances in which that melody is given to a piccolo, so one could reasonably speculate that Bartók thought of it as piped melody all along since it lies in such a high register on the piano. You can hear that version, starting around 1:55 here:



However, the radio performance I heard left me kind of cold because the version I really know and love is the violin/piano arrangement made by Zoltan Szekely - I've accompanied this version many times, including with my favorite home-grown violinist. Szekely has the violinist use artificial harmonics, presumably to imitate the glassy, pure sound of the piccolo or flute or pipe or whatever - but, of course, it doesn't really sound convincingly like any of those. What it sounds like (at 1:45) is eerie and haunting and odd and frail....and beautiful:



So, this pale imitation of a flute is by far my favorite way to hear this dance, even though the arrangement is just an imitation of the composer's orchestration. Imitation yields something new. Of course, personal bias plays a role, given that I've heard it most often this way, but I still think there's an argument that this is the most compelling version, and that's significantly because of its failings. It's a violin trying to sound like an orchestral flutist who presumably is trying to sound like a Romanian shepherd, and so it's a violin doing things the violin wasn't really even designed to do, using the artificial harmonics to take much of the body out of the sound and leave behind that ghostly trail.

Actually, I could easily imagine myself having made an argument on behalf of the piano version if that had been the one I'd known best, and though it's not my favorite here (the piano strikes me as too neutral), the truth is that, like most pianists, I have a very deep affection for all the ways in which the sound of a piano is weak and unevenly balanced and compromisingly tuned and poorly sustained. For all the great innovations that made the modern piano sound what it is, it is still a sonority sustained by smoke and mirrors, and we love it for its frailness, its vulnerability, its Walter Mitty-like attempts to imitate flutes and violins, horns and voices. We strive to make it sing with beautiful legato, but we don't really want it to do that perfectly. In fact, the bell-like tones of a Rhodes-style electric piano are the last thing most pianists want to hear when striking the keys to sing a Chopin nocturne. Modern synthesizer technology can fix a lot of the problems I've outlined, and many musical styles have taken advantage of that - but a "piano" can't be imitated; only the real thing fails in such a perfect way.

A favorite example is the piano opening of Strauss's incomparably perfect Morgen:



For some misguided reason (probably a singer who couldn't stand having to follow that perfect piano intro), Strauss decided to orchestrate this and give the halting, suspended piano melody to a sugary sweet violin and....well, here you go:



Sure, it's pretty, and if I didn't know the piano version, I might even love it, but...no. Pretty much every tune Chopin wrote falls into this category - tunes inspired by bel canto opera, but which sound most inspired when hammered on string.

---

There are many other ways to explore this question of imitation as improvement. I haven't yet played organ long enough to fall in love with all the bizarre organ stops that are supposedly trying to sound like flutes, and reeds, and trumpets, and violas, but apparently one can develop an actual affection for all of those imitations. I also was recently engaged in a Facebook discussion about computers composing music (inspired by this article), and when one commenter suggested that imitation can never replace inspiration, I openly wondered if inspiration might not sometimes result from mere imitation. Bach certainly got a lot of mileage out of copying Vivaldi. Or, to go in a different direction, one might note that, when going to see Shakespeare, we much prefer to see an imitation of someone being stabbed than the real thing.

But I'll leave those paths untaken and close with another bit of violin rep that's been on my mind. Daughter of MMmusing is now learning Ravel's great Tzigane, and we read it together for the first time last night. Here we have perhaps the most polished musical craftsman of all time creating a sort of imitation of a gypsy jam session (with the original piano part written for a piano that had an attachment designed to imitate the cimbalon!) - except, it's got some pretty meticulous stuff in there, and it's not really an invitation for the performers to unleash themselves with total abandon, at least not unless total abandon includes playing the tricky notes Ravel took the trouble to write down. Oh sure, there's still plenty of room for interpretive abandon, but as I've said about The Rite of Spring so many times, it's an imitation of the exotic/primitive that requires a very disciplined, high-culture brand of performance. That tension is at the heart of why music (and art) can be so compelling.

This isn't the place to argue whether such exotica is always better than the thing it's imitating (I heard one of the Liszt orchestral Hungarian Rhapsodies the other day and it kind of made me cringe), but I do think there's something extraordinarily satisfying about Ravel's fun-house imitation of the gypsy fiddler, even if it sounds no more like an authentic gypsy than Bartók/Szekely's violin sounds like a flute. The point is that the failure to imitate accurately may be the inspiration for something incomparably great.

---

UPDATE FROM YEARS LATER (11/12/14): I should have mentioned here that Jeremy Denk once complained that Tzigane is, in fact, too "clean":

Just the other day I was playing through Tzigane with Josh, in a rehearsal, and it was all a great deal of fun, and Josh sounded fabulous of course, and I was annoyed that I didn’t sound so fabulous in that annoying passage with the repeated notes … but I was thinking “it’s good, but it’s no Charles Ives.” Even the “dirty” gypsy notes in that piece sound clean, organized, shiny; everything is polished, glittering, sparkling, lush, perfectly voiced: sanitized? It smelt of PineSol, if PineSol were French. But not with Ives; he captures the Down & Dirty better than almost anyone. If he errs, he errs on the Dirty side; but his dirt is not vulgar, it is transcendental fertile earth with lots of terrific spiritual manure. Perhaps the hyper-cleanliness of Ravel is somewhat vulgar, in comparison with the honest, sprawling dirtiness of Ives? … at least that’s the way I feel. Bring on the hate mail!